A self-adaptive control based morphing wing flutter suppression stabilization method

Abstract

The application discloses a variable wing flutter suppression and stabilization method based on adaptive control, and relates to the technical field of aircraft aeroelastic active control. The application constructs a fault-tolerant control system for real-time diagnosis, online reconstruction and dynamic redistribution, continuously monitors the health of an actuator cluster, actively diagnoses the operation state of each actuator, identifies abnormal states such as saturation, partial failure or complete failure of the actuator, starts an online reconstruction mechanism, integrates fault information into real-time calculation of a control law, dynamically adjusts the feedback gain matrix of the controller, calculates a virtual control quantity of the reconstruction control law as a whole instruction, reanalyzes and distributes the virtual control quantity to all healthy actuators, and equivalently replaces the function of the original faulty actuator in aerodynamic effect through a cooperative mode such as differential combination, so that the integrity of the function is maintained, and the flutter suppression and stabilization control under actuator failure is realized.

Description

Technical Field

[0001] This invention relates to the field of active aeroelastic control technology for aircraft, specifically a flutter suppression and stabilization method for deformable wings based on adaptive control. Background Technology

[0002] Deformable wings adapt to different flight conditions by changing their aerodynamic shape, thereby improving full envelope performance. However, such flexible structures are more prone to aeroelastic flutter under certain flight conditions, a dynamic instability phenomenon that can lead to catastrophic consequences. Existing technologies typically employ active control methods, driving actuators arranged on the wing to suppress the vibration.

[0003] However, existing technologies have a key drawback: their control strategies heavily rely on the premise that all actuators are working properly. When one or more actuators experience performance degradation, saturation, or complete failure due to electrical faults, mechanical jamming, or physical damage, traditional fixed control laws not only experience a sharp decline in vibration suppression efficiency but may also introduce new instability factors due to the asymmetrical distribution of control forces. Currently, there is a lack of an effective method that can sense the health status of actuators in real time during flight and automatically and intelligently reallocate control resources to maintain system stability.

[0004] Therefore, there is an urgent need for a flutter suppression system with inherent fault tolerance, which can quickly reconfigure the control strategy to ensure flight safety when actuators fail. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a flutter suppression and stabilization method for deformable wings based on adaptive control. This method constructs a fault-tolerant control system with real-time diagnosis, online reconfiguration, and dynamic redistribution. It continuously monitors the health of the actuator cluster, actively diagnoses the operating status of each actuator, and identifies abnormal states such as actuator saturation, partial failure, or complete failure. It then initiates an online reconfiguration mechanism, integrating fault information into the real-time calculation of the control law. By dynamically adjusting the controller's feedback gain matrix, the virtual control quantity calculated by the reconfigured control law is treated as a whole instruction, re-analyzed, and redistributed to all healthy actuators. Through differential combination and other cooperative modes, the aerodynamic effect is equivalent to replacing the function of the original faulty actuator, thus maintaining functional integrity and achieving flutter suppression and stabilization control under actuator failure, thereby improving system survivability and mission reliability.

[0006] To solve the above-mentioned technical problems, this invention provides the following technical solution: a flutter suppression and stabilization method for deformable wings based on adaptive control, the specific steps of which are as follows:

[0007] S1. Real-time Monitoring and Health Diagnosis: Continuously monitors the operating parameters of each deformable actuator on the wing and diagnoses the health status of each actuator in real time. The health status includes normal, saturated, partially ineffective, and completely ineffective;

[0008] S2. Fault Triggering and Online Reconfiguration: Once any actuator is diagnosed to be in one of the three abnormal health states of saturation, partial failure, or complete failure, the control efficiency matrix of the system is reconfigured, and the online reconfiguration process of the adaptive control law is immediately triggered.

[0009] S3, Dynamic reallocation of control quantity: Using a reallocation algorithm, the desired virtual control quantity generated by S2 is allocated to all remaining healthy actuators;

[0010] S4, Fault-tolerant Cooperative Execution: The remaining healthy actuators receive the control commands reallocated by S3 and perform cooperative actions by changing their own deformation modes to generate an equivalent flutter suppression force.

[0011] S5. Stability Maintenance: The aircraft's response after vibration suppression is fed back through vibration sensors on the wings and compared with the state without vibration.

[0012] Furthermore, in the S1 real-time monitoring and health diagnosis, during each control cycle... Real-time data collection The operating parameters of an actuator include the input command voltage. Output displacement Drive current and preset displacement Calculate the absolute value of the displacement residual. And diagnose the actuator's health status in real time based on the actuator's operating parameters:

[0013] normal: And the absolute value of the drive current ,in The first displacement residual threshold is preset. This refers to the rated maximum current of the actuator.

[0014] saturation: or ,at the same time ,in and These represent the positive and negative physical travel limits of the actuator. The current threshold for saturation;

[0015] Partial failure: Furthermore, the actuator was not determined to be in a saturated state, in which The second displacement residual threshold is used to distinguish between partial failure and complete failure. ;

[0016] Completely ineffective: Or at command voltage Under normal circumstances, , ;

[0017] Based on the judgment results, for each actuator Assign a control cycle The only health status indicator in the country and .

[0018] Furthermore, the reconstructed control efficiency matrix The specific method is as follows:

[0019] Based on the initial control efficiency matrix ,in It is the system modal number. It is the total number of actuators. Each column Representing the The control efficiency of an actuator for different modes;

[0020] Based on the health status identifier obtained from S1 Reconstruct according to the following rules :

[0021] when If so, then keep that column, that is ;

[0022] when If so, then the elements in that column are set to zero. ;

[0023] when Then multiply the elements of that column by a decay factor. ,Right now ,in and The first The output displacement and preset displacement of each actuator.

[0024] Furthermore, in the S2 fault triggering and online reconfiguration, the specific steps of the online reconfiguration process are as follows:

[0025] Based on the health status diagnosed by S1, and the system matrix A under the current flight state and the reconstructed... Solve the Lyapunov equations Obtain the positive definite matrix ;

[0026] Using the matrix obtained by solving The reconstructed adaptive control law is generated, and its output virtual control quantity is: ,in, The virtual control quantity generated after reconstruction is a vector representing the total control effort required to suppress chatter. It is the state vector of the actuator, including the wing's bending displacement, torsional displacement, and rate of change. This is the reconstructed control efficiency matrix. This represents the matrix transpose operation. It is a positive definite control weight matrix used to balance control effectiveness and energy consumption.

[0027] Furthermore, in the dynamic reallocation of the S3 control quantity, the reallocation algorithm is used to map the virtual control quantity to the actuator space where the health state is normal, constructing an optimization problem with allocation accuracy and actuator constraints as objectives:

[0028] Target: ;

[0029] constraint: ;

[0030] in, Let be the normal actuator command vector to be determined. and These are the upper and lower limits of the physical output;

[0031] The optimization problem is solved using a numerical optimization algorithm to obtain the allocated instruction vector. .

[0032] Furthermore, the numerical optimization algorithm through Solve the optimization problem, where, It is the reconstructed control efficiency matrix The pseudo-reverse and .

[0033] Furthermore, the specific steps of the S4 fault-tolerant collaborative execution are as follows:

[0034] The allocated instruction vector output by S3 This is analyzed as the timing control signals for each normal actuator;

[0035] All normal actuators operate synchronously based on the analyzed signals. That is, when an actuator fails, its multiple adjacent actuators work together through a differential combination mode to synthesize the required control torque. The differential combination mode is used to drive multiple adjacent healthy actuators after the main actuator of the wing fails, causing them to deflect in different magnitudes and directions. The asymmetric local aerodynamic forces caused by the deflection are combined into an equivalent total control force and total control torque acting on the wing to replace the function of the original failed actuator.

[0036] Furthermore, the specific steps for maintaining stability in S5 are as follows:

[0037] Multiple vibration sensors deployed on the wing continuously collect the wing's dynamic response signals after S4 fault-tolerant collaborative execution. These vibration sensors include strain gauges and accelerometers. The dynamic response signals directly characterize the wing's state vector. ;

[0038] The collected state vector By comparing with the state without vibration, the state error vector is calculated;

[0039] The state error vector is used as the input to the adaptive control law after S2 reconstruction for dynamic redistribution of control quantity and fault-tolerant collaborative execution until convergence to the preset stable state.

[0040] Compared with existing technologies, this adaptive control-based flutter suppression and stabilization method for deformable wings has the following advantages:

[0041] This invention constructs a fault-tolerant control system with real-time diagnosis, online reconfiguration, and dynamic redistribution. It continuously monitors the health of the actuator cluster, proactively diagnoses the operating status of each actuator, and initiates an online reconfiguration mechanism by identifying abnormal states such as actuator saturation, partial failure, or complete failure. The fault information is integrated into the real-time calculation of the control law. By dynamically adjusting the feedback gain matrix of the controller, the virtual control quantity calculated by the reconfigured control law is treated as a whole instruction, re-analyzed, and redistributed to all healthy actuators. Through cooperative modes such as differential combination, the function of the original faulty actuator is equivalently replaced in terms of aerodynamic effect, thereby maintaining functional integrity and achieving stable control to suppress flutter under actuator failure, thus improving system survivability and mission reliability.

[0042] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0044] Figure 1 This is a flowchart of a flutter suppression and stabilization method for deformable wings based on adaptive control;

[0045] Figure 2 This is a flowchart illustrating the steps of a flutter suppression and stabilization method for a deformable wing based on adaptive control. Detailed Implementation

[0046] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0047] Example 1

[0048] This embodiment provides the working principle and detailed execution flow of a flutter suppression and stabilization method for deformable wings based on adaptive control. It aims to address the problem that when deformable wing actuators experience saturation, partial failure, or complete failure, the damping effectiveness of fixed control laws decreases, and even new instability factors arise. Figure 2 As shown, S1 continuously monitors the operating parameters of each deformable actuator on the wing and diagnoses the health status of each actuator in real time. The health states include normal, saturated, partially failed, and completely failed. Once S2 diagnoses any actuator in any of the three abnormal health states of saturation, partial failure, or complete failure, it reconstructs the system's control efficiency matrix and immediately triggers the online reconstruction process of the adaptive control law. S3 uses a redistribution algorithm to distribute the desired virtual control quantity generated by S2 to all remaining healthy actuators. S4 The remaining healthy actuators receive the control commands redistributed by S3 and perform coordinated actions by changing their own deformation modes to generate an equivalent flutter suppression force. S5 The vibration sensor on the wing provides feedback on the body response after vibration suppression and compares it with the vibration-free state to achieve flutter suppression and stable control under actuator failure, ensuring the aeroelastic stability and flight safety of the deformable wing under complex flight conditions.

[0049] (I) Real-time monitoring and health diagnosis phase (S1)

[0050] This phase continuously collects the operating parameters of all deformable actuators on the deformable wing and diagnoses the health status of each actuator in real time based on preset judgment rules, providing accurate basis for subsequent fault handling. In each control cycle... Inside, firstly, on the wing... Key operating parameters of each actuator are collected synchronously. , The total number of actuators; the collected parameters include the input command voltage. Actual output displacement Drive current Simultaneously, the preset displacement of the actuator within the current control cycle is retrieved. Based on the collected parameters, the first step is to calculate the absolute value of the displacement residual. The calculation formula is: This indicator measures the deviation between the actual displacement and the expected displacement of the actuator. It is one of the core bases for judging whether the actuator has performance abnormalities. Combining displacement residual and drive current, the actuator status is diagnosed one by one according to the following four categories of health status judgment rules:

[0051] Normal state determination: When the absolute value of the displacement residual satisfies , The first displacement residual threshold is preset, and the absolute value of the driving current satisfies... hour, The rated maximum current of the actuator represents the upper limit of the current when the actuator is operating normally, and is used to determine the actuator's current during the control cycle. The internal components are in a normal state. At this time, the actuator's output displacement can follow the preset displacement, the drive current does not exceed the rated range, and it can normally participate in chatter suppression control.

[0052] Saturation state determination: When the actual output displacement of the actuator reaches the physical stroke limit, that is... or , and These are the positive and negative physical travel limits of the actuator, respectively, and simultaneously satisfy the absolute value of the drive current. , If the saturation current threshold is reached, the actuator is determined to be in a saturated state. At this time, the actuator cannot continue to produce a larger displacement according to the command. Even if the input voltage is increased, the output displacement will no longer change, and the current will continue to be in an overload state. Its normal control function needs to be suspended.

[0053] Partial failure condition determination: When the absolute value of the displacement residual satisfies , The second displacement residual threshold is preset and This is used to distinguish between partial failure and complete failure. If the actuator is determined not to be in a saturated state as previously stated, it is determined to be in a partial failure state. At this time, there is a significant deviation between the output displacement of the actuator and the preset displacement, but it can still produce a certain displacement response, indicating that its control effectiveness has been reduced but not completely lost. Its effectiveness needs to be corrected in subsequent control.

[0054] Complete failure condition determination: A condition is considered a complete failure if one of the following two conditions is met: (1) the absolute value of the displacement residual... This indicates that the actual displacement of the actuator deviates greatly from the expected displacement, making it impossible to achieve basic control functions; secondly, the input command voltage... Normal, that is Within the preset normal voltage range, excluding cases where the actuator fails to respond due to abnormal commands, the drive current... And output displacement This indicates that the actuator may have completely lost its driving ability due to faults such as electrical circuit break or mechanical jamming, and is unable to produce any displacement movement.

[0055] Finally, for each actuator Assign the current control cycle The only health status indicator in the country ,and Normal, saturated, partially failed, completely failed It also aggregates and stores the health status indicators of all actuators, providing data support for the next stage of fault triggering and reconfiguration.

[0056] (II) Fault Triggering and Online Reconfiguration Phase (S2)

[0057] In this stage, when an abnormal state of the actuator is diagnosed, the system's control efficiency matrix is ​​quickly reconstructed, and a new adaptive control law is generated based on the reconstructed matrix to ensure that the control strategy can adapt to the fault condition. First, a fault trigger judgment is performed, and the health status of all actuators summarized in stage S1 is checked. Perform a traversal check; if any actuator exists... If any of the three abnormal states—"saturation," "partial failure," or "complete failure"—is detected, the online reconfiguration process of the adaptive control law is immediately triggered. If all actuators are in normal condition, the original control strategy is maintained, and no reconfiguration is required. The first step in the reconfiguration process is to reconstruct the system's control efficiency matrix. The specific steps are as follows:

[0058] Retrieve the initial control efficiency matrix of the system ,in The system modal number represents the number of vibration modes involved in wing flutter, such as bending modes, torsional modes, etc. The total number of actuators, matrix Each column Representing the The control effectiveness of an actuator for different system modes, i.e. The elements in the equation correspond to the actuators of the 1st, 2nd, ... th order, respectively. Control force coefficients of the first mode, initial matrix It is pre-calibrated by the aeroelastic properties test of the wing.

[0059] when In "normal" mode, the actuator's control performance remains unaffected, and the corresponding columns in the initial matrix are retained. ;

[0060] when When the actuator is "saturated" or "completely failed," it can no longer provide effective control force, and all elements in the corresponding column of the initial matrix are set to zero. This indicates that its control effectiveness across all system modes has failed;

[0061] when When there is a "partial failure", the control effectiveness of the actuator is reduced, and a reduction factor needs to be introduced. Correct the initial column vector, with a decay factor. The calculation formula is: ,in This represents the actual output displacement of the actuator. For the preset displacement, The range of values ​​is This directly reflects the ratio of the actual performance to the expected performance of the actuator. The corrected column vector is... This enables precise quantitative correction of the control performance of partially failed actuators.

[0062] The second step in the reconstruction process is to solve the Lyapunov equations to obtain the positive definite matrix P and generate the reconstructed adaptive control law:

[0063] Retrieve the system matrix under the current flight status System matrix Determined by the wing's mass, stiffness, damping characteristics, and aerodynamic parameters such as current flight speed and altitude, it reflects the system's dynamic characteristics and is combined with the reconstructed control efficiency matrix. Construct the Lyapunov equations: ,in, Representation matrix transpose, The control weight matrix is ​​positive definite. The larger the element, the heavier the penalty for the control quantity, which can prevent the actuator from consuming excessive energy. The state weight matrix is ​​positive definite and is used to measure the importance of system state deviations. The larger the element in the equation, the higher the requirement for state stability. The positive definite matrix is ​​obtained by solving the equation using numerical methods. ,matrix Its function is to ensure that the reconstructed control law enables the system to satisfy the Lyapunov stability condition, that is, the system state can converge to a stable point.

[0064] Based on the positive definite matrix obtained by the solution The reconstructed adaptive control law is generated, and its output is the desired virtual control quantity. The calculation formula is: ,in, This is the state vector of the actuator, containing the wing's bending displacement, torsional displacement, and their first derivatives, directly reflecting the wing's current vibration state. It is in vector form, and its dimension is equal to the total number of actuators. Consistent, each element represents the "control effort" required by the corresponding actuator to suppress flutter, which is the core basis for subsequent control quantity allocation.

[0065] (III) Dynamic reallocation stage of control quantity (S3)

[0066] The core of this stage is to process the desired virtual control quantity generated in stage S2. By optimizing the algorithm and mapping the actuators to a "normal" health state, an equivalent vibration suppression effect can still be achieved through the coordinated action of healthy actuators, even after excluding faulty actuators. First, the redistribution objective and constraints are defined, and the optimization problem is constructed:

[0067] Optimization objective: Minimize the deviation between the actual control force generated by the healthy actuator and the desired virtual control quantity, i.e., minimize the norm. ,in, The vector of normal actuator commands to be solved has the same dimension as the number of healthy actuators. Each element represents the actual control command assigned to the corresponding healthy actuator, such as voltage or displacement command. This objective ensures that the coordinated action of the healthy actuators can be as close as possible to the desired vibration damping effect.

[0068] Constraints: Considering the physical performance limitations of the actuator, the actual control commands for a healthy actuator must satisfy the following: ,in, and These are the upper and lower limits of the physical output of the healthy actuator, to prevent new malfunctions caused by commands exceeding the actuator's capabilities.

[0069] Next, a numerical optimization algorithm is used to solve the above optimization problem. In this embodiment, a solution method based on pseudo-inverse is used, and the specific formula is as follows: ,in, For the reconstructed control efficiency matrix The pseudo-reverse and The advantage of the pseudo-inverse algorithm lies in its ability to stably solve for the optimization objective and constraints even when faulty actuators cause linear dependence in the column vectors. To ensure the feasibility and effectiveness of control quantity allocation, the solution is obtained. Then, it is necessary to further verify whether it meets the requirements. If any individual element exceeds the constraint range, that element will be truncated. Partial take lower than Partial take Ultimately, a control command vector that conforms to the physical capabilities of the actuator is obtained.

[0070] (iv) Fault-tolerant and collaborative execution phase (S4)

[0071] The core of this stage is to process the control command vectors allocated in stage S3. The specific actions of a healthy actuator are transformed into the combined actions of differential combinations and other cooperative modes to synthesize equivalent flutter suppression forces and torques, thus replacing the function of the faulty actuator. Each element in the code, i.e., the control instruction assigned to a single healthy actuator, is parsed into a timing-based control signal. The timing information includes the instruction's start time, duration, rate of change, etc. For example, if... If an element in the sequence represents a displacement command, that displacement value must be converted into a target displacement sequence for the actuator at different times to ensure that the actuator can smoothly and accurately follow the command. All healthy actuators start their actions synchronously based on the parsed timing control signals. For actuators that cannot operate due to faults, a differential combination mode is used to achieve functional replacement. The specific process is as follows:

[0072] Identify the installation location and original control function of the faulty actuator;

[0073] Multiple healthy actuators adjacent to the faulty actuator are selected. By controlling these healthy actuators to generate deflection actions of different magnitudes and directions, these deflection actions will induce asymmetric local aerodynamic forces on the wing surface. Through aerodynamic coupling effects, these asymmetric aerodynamic forces will combine into a total control force and total control torque equivalent to the original faulty actuator, which will act on the flutter-sensitive area of ​​the wing. For example, when a certain main actuator of the wing is responsible for providing a downward torsional torque to suppress flutter, if the actuator fails completely, the adjacent healthy actuator on its left can be controlled to deflect upward and the adjacent healthy actuator on its right can be controlled to deflect downward. Through the asymmetric aerodynamic force difference between the left and right sides, a downward equivalent torsional torque is synthesized, thereby replacing the function of the original faulty actuator.

[0074] During the coordinated action execution, it is necessary to ensure that the timing of the actions of all healthy actuators remains synchronized to avoid deviations in the resultant torque due to action delays, which would affect the vibration suppression effect. Simultaneously, the actual displacement is fed back in real time through the position sensors built into the actuators. If any deviation is detected in the action of a healthy actuator, its control signal is fine-tuned promptly to ensure the accuracy of the coordinated action.

[0075] (v) Stable Maintenance Phase (S5)

[0076] The core of this stage is to use vibration sensors to provide feedback on the wing's vibration suppression effect, comparing it with a vibration-free stable state to form a closed-loop control. The control strategy is continuously adjusted until the system converges to a preset stable state. Multiple vibration sensors are deployed in the flutter-sensitive areas of the wing, including strain gauges to collect strain signals from the wing structure, reflecting the degree of bending or torsional deformation, and accelerometers to collect acceleration signals from the wing, reflecting vibration intensity. These sensors continuously collect the wing's dynamic response signals after the fault-tolerant collaborative execution in stage S4, and convert these signals into a system state vector. , The dimensions and S2 stage Consistent, including the wing's bending displacement, torsional displacement, and their rate of change, directly characterizes the wing's current vibration state, and will collect... By comparing the state error vector with the reference vector of the vibration-free stable state, the magnitude of the state error vector is calculated, which directly reflects the current vibration suppression effect of the wing. The state error vector is used as the input of the adaptive control law after the reconstruction in stage S2, and stage S2 is re-executed to adjust the control efficiency matrix. With adaptive control law, S3 is based on new The process of reallocating control variables and executing new coordinated actions by the S4 health actuator forms a closed-loop control. Through continuous acquisition, comparison, adjustment, and execution iteration, the amplitude of the state error vector is gradually reduced. At this point, the flutter of the wing is effectively suppressed and maintained in a preset stable state. The closed-loop control process stops or switches to a low-frequency monitoring mode to save system energy.

[0077] In summary, this embodiment fully implements a flutter suppression and stabilization method for deformable wings based on adaptive control. The entire implementation process effectively solves the limitations of fixed control laws under actuator failure. It can quickly respond and maintain the aeroelastic stability of deformable wings when actuators saturate, partially fail, or completely fail, significantly improving the flight safety and mission reliability of deformable wings. It provides a feasible implementation path for the engineering application of aeroelastic active control technology for aircraft.

[0078] Example 2

[0079] Based on Example 1, this example provides a specific step in the flutter suppression of a deformable wing based on adaptive control, as follows: Figure 1 As shown, the specific steps are as follows:

[0080] (1) Real-time monitoring and health diagnosis

[0081] Signal acquisition: In each control cycle, the operating parameters of each wing deformation actuator are acquired in real time, including input command voltage, actual output displacement, drive current and preset displacement.

[0082] Calculate residuals: Calculate the absolute value of the difference between the actual displacement and the preset displacement of each actuator, i.e., the displacement residuals.

[0083] Status determination: The collected parameters and calculated residuals are compared with a series of preset thresholds, and the health status is determined according to the established logic.

[0084] Output identifier: Assign a unique, real-time updated health status identifier to each actuator.

[0085] (2) Fault Triggering and System Reconfiguration

[0086] Fault Trigger: Continuously monitor the health status indicators of all actuators. Once any actuator is detected to be in an abnormal state, the fault-tolerant control process is immediately triggered.

[0087] Reconstruct the control efficiency matrix: Dynamically reconstruct the system's control efficiency matrix based on the health status indicators of all actuators.

[0088] Update the control law: Based on the reconstructed control efficiency matrix and the current flight state, recalculate the core parameters of the adaptive control law and generate a new control law suitable for the current working mode.

[0089] (3) Dynamic reallocation of control quantities

[0090] Constructing the allocation problem: Using the virtual control quantity calculated by the new control law as the objective and considering the physical output constraints of all healthy actuators, construct an optimization problem.

[0091] Solving the allocation instructions: The numerical optimization algorithm is used to solve the optimization problem and calculate the specific physical control instructions that each healthy actuator should execute.

[0092] (4) Fault-tolerant collaborative execution

[0093] Command distribution: The obtained physical control command vector is parsed and distributed to each healthy actuator.

[0094] Cooperative actuation: After receiving the command, all healthy actuators act synchronously and work together through differential combination mode to synthesize the damping force and torque equivalent to those before the failure in terms of aerodynamic effect, thus replacing the function of the failed actuator.

[0095] (5) Closed-loop stability maintenance

[0096] Effect perception: Vibration sensors distributed on the wings are used to collect the wing vibration response in real time after the vibration suppression operation.

[0097] Feedback and loop: The perceived actual vibration state is compared with the expected stable state to obtain the error. This error signal is fed back to the update control law loop of (2) as the input of the next control cycle, thus forming a closed loop to continuously suppress flutter until the wing recovers and maintains a stable state.

[0098] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A flutter suppression and stabilization method for deformable airfoils based on adaptive control, characterized in that, The specific steps of this method are as follows: S1. Real-time Monitoring and Health Diagnosis: Continuously monitors the operating parameters of each deformable actuator on the wing and diagnoses the health status of each actuator in real time. The health status includes normal, saturated, partially ineffective, and completely ineffective; S2. Fault Triggering and Online Reconfiguration: Once any actuator is diagnosed to be in one of the three abnormal health states of saturation, partial failure, or complete failure, the control efficiency matrix of the system is reconfigured, and the online reconfiguration process of the adaptive control law is immediately triggered. The control efficiency matrix of the reconfiguration system The specific method is as follows: Based on the initial control efficiency matrix ,in It is the system modal number. It is the total number of actuators. Each column Representing the The control efficiency of an actuator for different modes; Based on the health status identifier obtained from S1 Reconstruct according to the following rules : when If so, then keep that column, that is ; when If so, then the elements in that column are set to zero. ; when Then multiply the elements of that column by a decay factor. ,Right now ,in and The first The output displacement and preset displacement of each actuator; In the S2 fault triggering and online reconfiguration, the specific steps of the online reconfiguration process are as follows: Based on the health status diagnosed by S1, and the system matrix A under the current flight state and the reconstructed... Solve the Lyapunov equations Obtain the positive definite matrix ; Using the matrix obtained by solving The reconstructed adaptive control law is generated, and its output virtual control quantity is: ,in, The virtual control variables generated after reconstruction. It is the state vector of the actuator, including the wing's bending displacement, torsional displacement, and rate of change. This is the reconstructed control efficiency matrix. This represents the matrix transpose operation. It is a positive definite control weight matrix used to balance control effectiveness and energy consumption; S3, Dynamic reallocation of control quantity: Using a reallocation algorithm, the desired virtual control quantity generated by S2 is allocated to all remaining healthy actuators; S4, Fault-tolerant Cooperative Execution: The remaining healthy actuators receive the control commands reallocated by S3 and perform cooperative actions by changing their own deformation modes to generate an equivalent flutter suppression force. S5. Stability Maintenance: The aircraft's response after vibration suppression is fed back through vibration sensors on the wings and compared with the state without vibration.

2. The flutter suppression and stabilization method for deformable wings based on adaptive control according to claim 1, characterized in that, In the S1 real-time monitoring and health diagnosis, during each control cycle Real-time data collection The operating parameters of an actuator include the input command voltage. Output displacement Drive current and preset displacement Calculate the absolute value of the displacement residual. And diagnose the actuator's health status in real time based on the actuator's operating parameters: normal: And the absolute value of the drive current ,in The first displacement residual threshold is preset. This refers to the rated maximum current of the actuator. saturation: or ,at the same time ,in and These represent the positive and negative physical travel limits of the actuator. The current threshold for saturation; Partial failure: Furthermore, the actuator was not determined to be in a saturated state, in which The second displacement residual threshold is used to distinguish between partial failure and complete failure. ; Completely ineffective: Or at command voltage Under normal circumstances, , ; Based on the judgment results, for each actuator Assign a control cycle The only health status indicator in the country and .

3. The flutter suppression and stabilization method for deformable wings based on adaptive control according to claim 1, characterized in that, In the dynamic reallocation of the S3 control quantity, the reallocation algorithm is used to map the virtual control quantity to the actuator space where the health state is normal, and to construct an optimization problem with the objectives of allocation accuracy and actuator constraints: Target: ; constraint: ; in, Let be the normal actuator command vector to be determined. and These are the upper and lower limits of the physical output; The optimization problem is solved using a numerical optimization algorithm to obtain the allocated instruction vector. .

4. The flutter suppression and stabilization method for deformable wings based on adaptive control according to claim 3, characterized in that, The numerical optimization algorithm is through Solve the optimization problem, where, It is the reconstructed control efficiency matrix The pseudo-reverse and .

5. The flutter suppression and stabilization method for deformable wings based on adaptive control according to claim 3, characterized in that, The specific steps of the S4 fault-tolerant collaborative execution are as follows: The allocated instruction vector output by S3 This is analyzed as the timing control signals for each normal actuator; All normal actuators operate synchronously based on the analyzed signals. That is, when an actuator fails, its multiple adjacent actuators work together through a differential combination mode to synthesize the required control torque. The differential combination mode is used to drive multiple adjacent healthy actuators after the main actuator of the wing fails, causing them to deflect in different magnitudes and directions. The asymmetric local aerodynamic forces caused by the deflection are combined into an equivalent total control force and total control torque acting on the wing to replace the function of the original failed actuator.

6. The flutter suppression and stabilization method for deformable wings based on adaptive control according to claim 1, characterized in that, The specific steps for maintaining stability in S5 are as follows: Multiple vibration sensors deployed on the wing continuously collect the wing's dynamic response signals after S4 fault-tolerant collaborative execution. These vibration sensors include strain gauges and accelerometers. The dynamic response signals directly characterize the wing's state vector. ; The collected state vector By comparing with the state without vibration, the state error vector is calculated; The state error vector is used as the input to the adaptive control law after S2 reconstruction for dynamic redistribution of control quantity and fault-tolerant collaborative execution until convergence to the preset stable state.

Images